feat(interactive): Import VARCHAR to GraphScope interactive (#3404)

Import a new data type `VARCHAR` to GraphScope Interactive. #3400 #3405 0. Use `VARCHAR` in schema. `VARCHAR` is not a primitive type. The following Representation can be revised. ```yaml property_type: var_char: max_length: 128 ``` 1. Refactor `PropertyType` Previous `PropertyType` is just a enum. ```c++ enum class PropertyType { kEmpty, kBool, kUInt8, kUInt16, ... }; ``` To represent `VARCHAR`, an enum alone is not enough, because we also need to express the maximum length of `VARCHAR`, i.e. max_length. ```c++ namespace impl { enum class PropertyTypeImpl { kEmpty, kBool, kUInt8, kUInt16, ... }; // Stores additional type information for PropertyTypeImpl union AdditionalTypeInfo { uint16_t max_length; // for varchar }; } // namespace impl struct PropertyType { impl::PropertyTypeImpl type_enum; impl::AdditionalTypeInfo additional_type_info; PropertyType() : type_enum(impl::PropertyTypeImpl::kEmpty), additional_type_info() {} PropertyType(impl::PropertyTypeImpl type) : type_enum(type), additional_type_info() {} PropertyType(impl::PropertyTypeImpl type, uint16_t max_length) : type_enum(type), additional_type_info({.max_length = max_length}) { assert(type == impl::PropertyTypeImpl::kVarChar); } //get DataType object, like `arrow::bool()`(but arrow returns a shared_ptr) static PropertyType empty(); static PropertyType bool_(); static PropertyType uint8(); static PropertyType uint16(); // different from other functions static PropertyType var_char(uint16_t max_length); //... static const PropertyType kEmpty; static const PropertyType kBool; static const PropertyType kUInt8; static const PropertyType kUInt16; //... bool operator==(const PropertyType& other) const; bool operator!=(const PropertyType& other) const; }; ``` VarChar is implemented but not fully provided by interactive. User shall be able to use VarChar type after VarChar is supported by compiler. --------- Co-authored-by: liulx20 <[email protected]>
alibaba · Dec 14, 2023 · 975667e · 975667e
1 parent 052b9e6
commit 975667e
Show file tree

Hide file tree

Showing 28 changed files with 663 additions and 445 deletions.
diff --git a/docs/flex/interactive/data_model.md b/docs/flex/interactive/data_model.md
@@ -67,8 +67,6 @@ property_type:
   primitive_type: DT_STRING
 ```
 
-Please note that we currently do not support the use of string data type for properties on edges.
-
 ### Array Types
 
 Array types are currently not supported, but are planned to be supported in the near future.

diff --git a/flex/bin/interactive_server.cc b/flex/bin/interactive_server.cc
@@ -314,6 +314,7 @@ int main(int argc, char** argv) {
     data_path = vm["data-path"].as<std::string>();
 
     auto schema = gs::Schema::LoadFromYaml(graph_schema_path);
+
     // Ths schema is loaded just to get the plugin dir and plugin list
     gs::init_codegen_proxy(vm, graph_schema_path, engine_config_file);
     db.Close();

diff --git a/flex/engines/graph_db/database/graph_db.cc b/flex/engines/graph_db/database/graph_db.cc
@@ -76,6 +76,7 @@ Result<bool> GraphDB::Open(const Schema& schema, const std::string& data_dir,
   }
 
   if ((!create_empty_graph) && (!graph_.schema().Equals(schema))) {
+    LOG(ERROR) << "Schema inconsistent..\n";
     return Result<bool>(StatusCode::InternalError,
                         "Schema of work directory is not compatible with the "
                         "graph schema",

diff --git a/flex/engines/graph_db/database/transaction_utils.h b/flex/engines/graph_db/database/transaction_utils.h
@@ -26,74 +26,53 @@
 namespace gs {
 
 inline void serialize_field(grape::InArchive& arc, const Any& prop) {
-  switch (prop.type) {
-  case PropertyType::kBool:
+  if (prop.type == PropertyType::Bool()) {
     arc << prop.value.b;
-    break;
-  case PropertyType::kInt32:
+  } else if (prop.type == PropertyType::Int32()) {
     arc << prop.value.i;
-    break;
-  case PropertyType::kUInt32:
+  } else if (prop.type == PropertyType::UInt32()) {
     arc << prop.value.ui;
-    break;
-  case PropertyType::kDate:
+  } else if (prop.type == PropertyType::Date()) {
     arc << prop.value.d.milli_second;
-    break;
-  case PropertyType::kString:
+  } else if (prop.type == PropertyType::String()) {
     arc << prop.value.s;
-    break;
-  case PropertyType::kEmpty:
-    break;
-  case PropertyType::kInt64:
+  } else if (prop.type == PropertyType::Int64()) {
     arc << prop.value.l;
-    break;
-  case PropertyType::kUInt64:
+  } else if (prop.type == PropertyType::UInt64()) {
     arc << prop.value.ul;
-    break;
-  case PropertyType::kDouble:
+  } else if (prop.type == PropertyType::Double()) {
     arc << prop.value.db;
-    break;
-  case PropertyType::kFloat:
+  } else if (prop.type == PropertyType::Float()) {
     arc << prop.value.f;
-    break;
-  default:
-    LOG(FATAL) << "Unexpected property type";
+  } else if (prop.type == PropertyType::Empty()) {
+  } else {
+    LOG(FATAL) << "Unexpected property type" << int(prop.type.type_enum);
   }
 }
 
 inline void deserialize_field(grape::OutArchive& arc, Any& prop) {
-  switch (prop.type) {
-  case PropertyType::kBool:
+  if (prop.type == PropertyType::Bool()) {
     arc >> prop.value.b;
-    break;
-  case PropertyType::kInt32:
+  } else if (prop.type == PropertyType::Int32()) {
     arc >> prop.value.i;
-    break;
-  case PropertyType::kUInt32:
+  } else if (prop.type == PropertyType::UInt32()) {
     arc >> prop.value.ui;
-    break;
-  case PropertyType::kDate:
+  } else if (prop.type == PropertyType::Date()) {
     arc >> prop.value.d.milli_second;
-    break;
-  case PropertyType::kString:
+  } else if (prop.type == PropertyType::String()) {
     arc >> prop.value.s;
-    break;
-  case PropertyType::kEmpty:
-    break;
-  case PropertyType::kInt64:
+  } else if (prop.type == PropertyType::Int64()) {
     arc >> prop.value.l;
-    break;
-  case PropertyType::kUInt64:
+  } else if (prop.type == PropertyType::UInt64()) {
     arc >> prop.value.ul;
-    break;
-  case PropertyType::kDouble:
+  } else if (prop.type == PropertyType::Double()) {
     arc >> prop.value.db;
-    break;
-  case PropertyType::kFloat:
+  } else if (prop.type == PropertyType::Float()) {
     arc >> prop.value.f;
-    break;
-  default:
-    LOG(FATAL) << "Unexpected property type";
+  } else if (prop.type == PropertyType::Empty()) {
+  } else {
+    LOG(FATAL) << "Unexpected property type: "
+               << static_cast<int>(prop.type.type_enum);
   }
 }
 

diff --git a/flex/engines/hqps_db/core/operator/sink.h b/flex/engines/hqps_db/core/operator/sink.h
@@ -179,41 +179,31 @@ void template_set_tuple_value(results::Collection* collection,
 /////Sink Any
 
 void set_any_to_common_value(const Any& any, common::Value* value) {
-  switch (any.type) {
-  case gs::PropertyType::kBool:
+  if (any.type == PropertyType::Bool()) {
     value->set_boolean(any.value.b);
-    break;
-  case gs::PropertyType::kInt32:
+  } else if (any.type == PropertyType::Int32()) {
     value->set_i32(any.value.i);
-    break;
-  case gs::PropertyType::kInt64:
-    value->set_i64(any.value.l);
-    break;
-  case gs::PropertyType::kUInt32:
-    // FIXME(zhanglei): temporarily use i64, fix this after common.proto is
+  } else if (any.type == PropertyType::UInt32()) {
+    // FIXME(zhanglei): temporarily use i32, fix this after common.proto is
     // changed
     value->set_i32(any.value.ui);
-    break;
-  case gs::PropertyType::kUInt64:
+  } else if (any.type == PropertyType::Int64()) {
+    value->set_i64(any.value.l);
+  } else if (any.type == PropertyType::UInt64()) {
     // FIXME(zhanglei): temporarily use i64, fix this after common.proto is
     // changed
     value->set_i64(any.value.ul);
-    break;
-  case gs::PropertyType::kDate:
-    value->set_i64(any.value.d.milli_second);
-    break;
-  case gs::PropertyType::kString:
-    value->mutable_str()->assign(any.value.s.data(), any.value.s.size());
-    break;
-  case gs::PropertyType::kDouble:
+  } else if (any.type == PropertyType::Double()) {
     value->set_f64(any.value.db);
-    break;
-  case gs::PropertyType::kFloat:
+  } else if (any.type == PropertyType::Float()) {
     value->set_f64(any.value.f);
-    break;
-  default:
-    LOG(WARNING) << "Unsupported type: " << any.type;
-    break;
+  } else if (any.type == PropertyType::Date()) {
+    value->set_i64(any.value.d.milli_second);
+  } else if (any.type == PropertyType::String()) {
+    value->mutable_str()->assign(any.value.s.data(), any.value.s.size());
+  } else {
+    LOG(WARNING) << "Unexpected property type: "
+                 << static_cast<int>(any.type.type_enum);
   }
 }
 

diff --git a/flex/engines/hqps_db/database/mutable_csr_interface.h b/flex/engines/hqps_db/database/mutable_csr_interface.h
@@ -852,7 +852,7 @@ class MutableCSRInterface {
           *std::dynamic_pointer_cast<TypedColumn<float>>(column));
     } else {
       LOG(FATAL) << "unexpected type to create column, "
-                 << static_cast<int>(type);
+                 << static_cast<int>(type.type_enum);
       return nullptr;
     }
   }

diff --git a/flex/engines/hqps_db/structures/multi_edge_set/untyped_edge_set.h b/flex/engines/hqps_db/structures/multi_edge_set/untyped_edge_set.h
@@ -248,7 +248,7 @@ class UnTypedEdgeSet {
     while (iter != end_iter) {
       auto edata = iter.GetData();
       PropT prop;
-      if (edata.type == TypeConverter<PropT>::property_type) {
+      if (edata.type == TypeConverter<PropT>::property_type()) {
         ConvertAny<PropT>::to(edata, prop);
       }
       CHECK(cur_ind < sum) << "cur: " << cur_ind << ", sum: " << sum;

diff --git a/flex/storages/rt_mutable_graph/dual_csr.h b/flex/storages/rt_mutable_graph/dual_csr.h
@@ -164,8 +164,10 @@ class DualCsr : public DualCsrBase {
 template <>
 class DualCsr<std::string_view> : public DualCsrBase {
  public:
-  DualCsr(EdgeStrategy oe_strategy, EdgeStrategy ie_strategy)
-      : in_csr_(nullptr), out_csr_(nullptr), column_(StorageStrategy::kMem) {
+  DualCsr(EdgeStrategy oe_strategy, EdgeStrategy ie_strategy, uint16_t width)
+      : in_csr_(nullptr),
+        out_csr_(nullptr),
+        column_(StorageStrategy::kMem, width) {
     if (ie_strategy == EdgeStrategy::kNone) {
       in_csr_ = new EmptyCsr<std::string_view>(column_, column_idx_);
     } else if (ie_strategy == EdgeStrategy::kMultiple) {